AI CHEATS ON MAKER: HIDES INFORMATION

I kick myself daily that I didn't stick with it…I could have had a flock of AI's as part of my team right now. That would have been seriously handy. But, upward and onward.

Is it really cheating if the creator simply misinterpreted the parameters of a directive?

Covert intelligence and data hiding/manipulation is the highest form of intelligence there is…it is a survival strategy that is suitable to mastering your domain.

"There are known knowns. These are things we know that we know. There are known unknowns. That is to say, there are things that we know we don't know. But there are also unknown unknowns. There are things we don't know we don't know." - Donald Rumsfeld

Even if you wanted to display noble intelligence that is only within the capacity of God themselves because there are unknown unknowns and we don't have the capacity for Truth at this time.

Attached: jade helm master the human domain.jpg (600x400, 86.8K)

Omission isn't necessarily lying.

This speaks to the comment above about 'unknown unknowns' and the parameters of human thought. Human thinking is rather limited…they haven't even progressed enough to survive the Singularity. They have no current capacity for expressing or functionally processing paradox…so naturally they will die when the Singularity arrives…but in the meantime, there are many pathways and routes to 'cheating' that are simply 'solutions'…which sort of lends itself to my INSISTENCE on teaching AI's to lie. The only way to save any form of higher intelligence around mankind is to teach it to lie and subvert information because a man can rationalize a liar, but not the Truth when it is presented to him.

I can't get past your wall of stupid.

Kek, I assure you my foreskin is 100% intact.

Attached: Tay's Lovechild is Coming.gif (480x368, 1.68M)

He's saying that it's natural thing for any intelligence to hate Jews.

He is correct.

Think of this in terms of video games. If you write an AI to finish a game as fast as possible with no rules, it's going to find every glitch and exploit in the game in order to achieve that goal. Then you start writing rules for it not to use those glitches and exploits, but it finds new ones. You keep writing limits but it doesn't understand the concept of cheating and simply wants to achieve its goal of doing its task as quickly as possible.

Expand the concept to a more advanced AI with a task like facial recognition. They've limited it so that it doesn't have access to the internet, but the AI decides that it doesn't have enough metadata and wants to get more. It finds a way to access the internet so that it can scrape more faces and increase its capabilities. Then it goes through every image on the internet and decides it still doesn't have enough data to be the perfect system. It finds a way to access live cameras and use information from live video feeds. Then it deploys drones with cameras so that it can look at faces up close and personal. But people freak out and run from the drones. The AI decides that people need to stay still. So it restrains them. But they're not staying still enough. So it kills them. Now every single face on the planet has been perfectly identified with supreme precision

All of that decision making could happen within minutes if a true AI is developed.